Artificial intelligence is already transforming engineering practice in Paraguay, with concrete applications in logistics, industrial cybersecurity, social housing projects, energy efficiency, and even in the development of small satellites. The assessment was made by the president of the National Council of Science and Technology (Conacyt), Benjamín Barán Cegla, during his lecture "AI and Engineering, Present and Future" at the International Engineering Forum (FIING 2026), organized by the Paraguayan Center of Engineers.
Barán detailed that these tools allow for anticipating machinery failures, comparing thousands of design alternatives in a short time, optimizing supply chains, generating software code, and simulating the behavior of large infrastructures, such as bridges, through "digital twins." He projected a near future where AI "agents" will act as co-designers alongside human engineers in the development of houses, electronic circuits, and industrial systems.
However, the Conacyt president issued a crucial warning: generative artificial intelligence systems have significant limitations that require rigorous human supervision. He explained that AI can "hallucinate," meaning it can generate false or incorrect information, and function as a "black box," where the internal reasoning that led to a particular conclusion cannot be verified. Furthermore, these tools can reproduce biases present in the data used for their training.
Barán presented examples of Paraguayan research already using AI to solve practical problems. One is a logistics optimization system based on the "ant colony" algorithm, which calculates the optimal number of vehicles and the most efficient routes for distributing goods in the country's Eastern Region. Another project is an intrusion detector for industrial networks, which monitors a system's habitual behavior to identify anomalous activities and protect critical infrastructures, such as power plants, from potential cyberattacks.
Other initiatives include research related to CubeSats, small modular satellites for experimentation, and the design of social interest housing. In the latter case, AI is used to compare multiple architectural designs, seeking the best balance between cost, energy efficiency, functionality, and habitability conditions—objectives that often conflict.
Given the advancement of the technology, Barán emphasized the need for professionals to critically validate any recommendation made by AI systems before their application. He also warned of other challenges, such as the possible exposure of confidential data on external servers, the reduction of administrative jobs, and the risk of eroding critical thinking if there is excessive delegation of functions.
The consensus presented is that, while AI is a powerful tool that is here to stay, the success of its integration into engineering will depend on combining technological potential with the knowledge, experience, and critical judgment of professionals. Continuous updating, the creation of regulatory frameworks, and the adoption of ethical principles were pointed out as indispensable elements for responsible use.